Understanding the molecular mechanisms used by human immunodeficiency virus (HIV)-1 to kill CD4+ T lymphocytes should contribute to finding new therapies capable of interrupting the progression to AIDS which follows HIV infection. Studies with peripheral blood T cells infected with primary clinical isolates of HIV-1 have shown that HIV-1 kills T cells by trapping the cells at a late point in the cell cycle (G2/M interface). These activated but arrested proliferating cells die when they are unable to complete the cell cycle and an abnormal form of programmed cell death (PCD) is initiated. The characteristics of this PCD, which include accumulation of proteins found only in cells about to enter mitosis, define the killing event as a pre-mitotic catastrophe. Working with cell lines expressing mutant forms of the cyclin dependent kinase p32cdc2 we were able to demonstrate that alterations in cdc2 kinase regulation triggers biochemical abnormalities that strongly resemble the biochemical abnormalities seen in the setting of HIV infection. Additionally, overexpression of bcl-2 was not seen in the setting of HIV infection and overexpression of bcl-2 was not able to block HIV-mediated PCD. We further demonstrated with a panel of metabolic inhibitors that HIV- mediated PCD substantially differed from the programmed death of CD4+ T cells which occurs during normal T-cell development (negative selection). These observations have thereby increased the likelihood that clinically useful agents can be developed to selectively inhibit the HIV-killing process without initiating unacceptable toxicity in normal T cells.